The Longitudinal Study of Australian Children Annual statistical report 2013

9 Fathers' involvement in the lives of their children: Separated parents' preferences

Lixia Qu and Ruth Weston, Australian Institute of Family Studies

Download printable version: Fathers' involvement in the lives of their children (PDF 630 KB)

9.1 Introduction

The progressive increase in the number of mothers in the labour force over the last few decades represents one of the greatest changes to have occurred in family life - not only in terms of the absolute size of the change, but also in terms of its effect on the way families function. In essence, the "male-breadwinner female-homemaker" model that was ubiquitous in the post-war boom period has given way to a shared (paternal and maternal) breadwinning role. Although various adjustments in the home, workplace and community to facilitate this new way of life have taken time to emerge, there is now evidence that fathers in Australia are spending more time caring for their children today than they were in the early 1990s (Craig, Mullan, & Blaxland, 2010). An increase in paternal involvement in the everyday lives of children - including time spent caring for the children - has been observed in other Western countries as well (see Moloney, Weston, & Hayes, 2013).

While their roles have been traditionally recognised as the economic providers in families, research suggests that fathers have a significant influence on their children's development (Lamb & Lewis, 2012). In general, relationships between fathers and children differ from those of mothers and children. For example, fathers are more likely than mothers to engage in physically stimulating activities with their children and to encourage independence (see Flouri, 2007; Lamb & Lewis, 2012). Some studies have suggested that fathers' active involvement with their children has beneficial spin-offs in relation to children's adjustment and wellbeing (for a review of relevant literature, see Allen & Daly, 2007). However, the quality of parenting is particularly important (see Allen & Daly, 2007; Marsiglio, Amato, Day, & Lamb, 2000). That is, investment of time is a necessary but not sufficient condition for high quality parenting. This finding has emerged not only in research on intact families, but also in research on families where parents have separated (see Allen & Daly, 2007). For instance, in their analysis of the findings of 63 studies focusing on separated fathers, Amato and Gilbreth (1999) found that children had better academic and socio-emotional outcomes when they felt close to their fathers and when their fathers adopted an authoritative parenting style.

The observed link between the level of paternal involvement and children's developmental progress may be because, as several studies have suggested, highly engaged fathers are more likely than other fathers to adopt high quality parenting, which, in turn, generates positive father-child relationships (see Allen & Daly, 2007).1 This is not to suggest, however, that all fathers who spend considerable time with their children are predisposed to engage with their children in beneficial ways.

Consistent with this body of research, the 2006 amendments to the Family Law Act 1975 emphasised, among other things, the importance of the continuing involvement of both parents in the lives of their children after parental separation, as long as this does not jeopardise their safety.2 These amendments are compatible with the views of Australian parents on the links between post-separation parental involvement and child wellbeing. In two separate national surveys of Australian parents with children under 18 years of age (conducted in 2006 and 2009), over three-quarters agreed with the statement that "children generally do best after separation when both parents stay involved in their lives" (Kaspiew et al., 2009). In addition, the level of support for this view appeared to have increased slightly between 2006 and 2009. Furthermore, separated fathers in both surveys were more likely than separated mothers to agree with this statement. This is not surprising, given that children of separated parents typically spend considerably more time in the care of their mothers than their fathers after parental separation, with a substantial proportion seeing their father less than once a year or never (see Australian Bureau of Statistics (ABS), 2011). The general consensus on the importance of the continuing involvement of both parents in their children's lives after parental separation (where this does not jeopardise children's safety) is consistent with the observed increase in the amount of time fathers spend with their children in Australia and some other Western countries (Craig et al., 2010; Moloney et al., 2013).

Caring for children on an everyday basis, supporting them financially, and making decisions that affect their long-term welfare represent key aspects of parental involvement, though not all these aspects may be considered when separated parents reflect on their preferences for their own and the other parents' involvement in their children's lives. Indeed, issues relating to care time may dominate their considerations, though some parents may focus more on financial provisions for their children.

Some studies have suggested that, where children of separated parents spend most or all nights in the care of mothers, most fathers would like to have increased involvement, while only a minority of mothers want to see this happen. For example, in a nationally representative Australian study, Parkinson and Smyth (2004) found that for children who spend most or all nights in the care of mothers, nearly 60% of fathers wanted to spend more time with their children, while more than half of the mothers were happy with the status quo.

Baxter, Edwards, and Maguire (2012) also examined patterns of preferences held by separated parents who are participating in the Longitudinal Study of Australian Children (LSAC). Their analysis focused on mothers who cared for the study child for most or all nights and fathers who were living elsewhere and saw the child at least once a year, specifically, both parents of the same study child who participated in the study in Wave 3 (here called the "paired cases sample"). While 75% of fathers in the paired cases sample indicated that they would prefer to have increased involvement with their child, only 45% of the mothers in this sample said they would prefer their child's father to have greater involvement than he currently had.

Separated parents may have misgivings about the increased involvement of the other parent for several reasons, including the very young age of their child, distance between the homes, and the perceived capacity of the other parent to be a good parent. Regarding the latter issue, for example, in the first wave of the Longitudinal Study of Separated Families, half of the mothers held safety concerns associated with the child's ongoing contact with the other parent (Kaspiew et al., 2009).3 In the second wave of this study, the capacity of the other parent to provide a safe environment for the children and to engage in appropriate parenting were common themes reported by those who held safety concerns (Qu & Weston, 2010). Such concerns were related with the quality of the inter-parental relationship, which is also a key factor in shaping children's wellbeing.

Expanding on previous research by Baxter, Edwards, and Maguire (2012), this chapter examines separated parents' preferences regarding the father's involvement in the life of their child using the data collected in four waves of LSAC. The chapter addresses the following issues:

  • What were separated mothers' and fathers' views regarding the involvement of the father in their child's life across four waves?
  • To what extent did the preferences change regarding the father's involvement?
  • To what extent did views between the mother and father in the "paired cases sample" differ regarding the father's involvement in their child's life?
  • What did fathers say in terms of barriers that prevent them from having more involvement?
  • What factors were linked with the views of the separated parents regarding the father's involvement?

9.2 Defining the sample of separated parents

The analysis in this chapter is based on the reports of separated mothers who usually lived with the study child (i.e., spend the most time with the child) and who had provided much of the information about this child, and on the reports of fathers who lived elsewhere from the mother.4 These parents are here called "resident mothers" and "non-resident fathers". The "study child" is also referred to as the "child". Note that the term "resident mother" is not intended to suggest that among these families, the mother-child relationship is more important than the father-child relationship.

It should be noted that although LSAC collected information from the parent who spent the most time with the study child in each wave, some parents had a shared care-time arrangement.5 In addition, some parents who participated in the study had never lived with their child's other parent. Mothers whose child had a shared care-time arrangement and mothers who had never lived with their child's father were also included in this analysis and are also referred to as "resident mothers", while the fathers are referred to as "non-resident fathers". These mothers lived with the child and said that they knew the child best.6

Omitted from the analysis are: (a) mothers who did not provide the majority of information about the child (representing fewer than 11 in B cohort mothers and six K cohort mothers across the study waves); and (b) fathers who usually lived with the study child (representing fewer than 15 B cohort fathers and fewer than 47 K cohort fathers across the study waves), given that they were not asked about their preferences for their own involvement in their children's lives. It is important to note that parents living elsewhere in Wave 1 were not interviewed. For ease of discussion, we refer to the mothers as the "resident mother" and fathers as the "non-resident father", as noted above.

Table 9.1 shows numbers of B cohort and K cohort resident mothers and non-resident fathers who were represented in the analysis. More resident mothers than non-resident fathers were represented, and the number of participants varied across waves. Some parents separated between waves and therefore "flowed into" the sample, while some "left" the sample, either because they no longer participated in the study or because they did not answer questions of prime interest in this analysis.7 As noted above, only fathers who had seen their child at least once a year were able to participate in the study. Of the mothers who had never lived with their child's father, 36% in the B cohort and 50% in the K cohort participated in all four waves, and 28% and 41% in the two cohorts respectively had responded to the question on their preferences regarding the father's involvement in their child's life. Of the non-resident fathers who participated in the study in Wave 2, 48% and 52% respectively participated in all subsequent waves and responded to the question regarding their own involvement.

Table 9.1: Number of resident mothers and non-resident fathers in each study wave, B and K cohorts
Wave and study year a B cohort K cohort
Age of child No. of resident mothers Non-resident fathers (interviewed or with self-completion questionnaire b Age of child No. of resident mothers Non-resident fathers (interviewed or with self-completion questionnaire b
Wave 1 (2004) 0-1 year 443 N/A 4-5 years 713 N/A
Wave 2 (2006) 2-3 years 475 91 6-7 years 682 183
Wave 3 (2008) 4-5 years 447 257 8-9 years 633 370
Wave 4 (2010) 6-7 years 631 355 10-11 years 781 440

Notes: a See Renda (2013) for detailed care-time arrangements in each wave. b Non-resident parents were sent self-completion questionnaires in Wave 2 and were interviewed via telephone in Waves 3 and 4.

A minority of resident mothers who were identified as the primary carer of the study child and non-resident fathers who were identified as the parent living elsewhere were in shared care-time arrangements. In Wave 3, 97 resident mothers of B cohort and 118 mothers of K cohort children did not want to answer questions about their child's other parent and thereby skipped answering the question on their preference regarding paternal involvement in their child's life. For other waves (1, 2 and 4), a small number of resident mothers who did not answer the question are excluded (ranging from 3 to 32 across waves for both cohorts). A small number of resident fathers living elsewhere who did not respond to the question about their own involvement in the study child's life are excluded (two or fewer for each wave and each cohort). In Waves 3 and 4, parents living elsewhere were interviewed via telephone.

9.3 Preferences of mothers and fathers regarding the father's involvement in their child's lives

In each wave, resident parents were asked: "How involved do you think [the child's] other parent should be in his/her life?" The response options were: "a lot more involved", "a little more involved", "the level of involvement is about right", "a little less involved", and "much less involved". As noted above, non-resident parents were asked about their preferred level of involvement in their child's life from Wave 2 onwards. These fathers were asked: "In thinking about the role that you have in this child's life, would you like to be … ?" The same set of response options were offered. The responses of resident mothers and non-resident fathers in relation to their views about the father's involvement in their children's lives are depicted in Figures 9.1 and 9.2. It should be pointed out that a father's involvement is multidimensional and parents' views may vary regarding the meaning of "father's involvement".

Figure 9.1: Preferences of resident mothers about the non-resident father's level of involvement in their child's life by study wave and cohort

Preferences of resident mothers about the non-resident father’s level of involvement in their child’s life by study wave and cohort

Note: Sample sizes of mothers across the four waves for each cohort vary. B cohort: n = 443, 475, 447 and 631 for Waves 1 to 4 respectively; K cohort: n = 713, 682, 633, and 781 for Waves 1 to 4 respectively.

Figure 9.2: Preferences of non-resident fathers about the level of their personal involvement in their child's life by study wave and cohort

Preferences of non-resident fathers about the level of their personal involvement in their child’s life by study wave and cohort

Note: Sample sizes of fathers across the four waves for each cohort vary. B cohort: n = 91, 257, and 355 for Waves 2 to 4 respectively; K cohort: n = 183, 370, and 440 for Waves 2 to 4 respectively.

Preferences for paternal involvement (from both mothers and fathers)

Around one-half of the resident mothers expressed a preference for increased paternal involvement (i.e., a lot more or little more involved). Across the waves of the study, nearly one-third of resident mothers in each cohort indicated that they would like the father of the child to be a lot more involved (B cohort: 29-33%; K cohort: 28-32%), and at least one-fifth said they preferred the father to be a little more involved (20-23% in each cohort) (Figure 9.1). On the other hand, a substantial proportion of resident mothers said that the current level of paternal involvement was "about right". This view was expressed by 38-41% of those in the B cohort and by 40-44% of K cohort resident mothers. Few mothers preferred to see the father's level of involvement diminish. Across study waves no more than 3% in each cohort said they preferred "a little less" involvement, and 3-9% in the B cohort and 5-8% in the K cohort said they preferred "much less" involvement.

The patterns of preferences were therefore fairly consistent across all the waves and the two cohorts. Nevertheless, though few resident mothers said that they would like the father to be less involved, this preference was selected by a significantly lower proportion of B cohort mothers in Wave 1 than in subsequent study waves (4% vs 8-13%). This trend was not apparent for K cohort mothers.

Figure 9.2 presents the patterns of preferences expressed by non-resident fathers concerning their personal involvement in their child's life. As noted above, such information was not collected in Wave 1 and the sample of non-resident fathers was restricted to those who had seen their child at least once a year. Fathers most commonly indicated that they preferred "a lot more" involvement (reported by 39-45% in the B cohort and 39-52% in the K cohort across Waves 2 to 4). Another 22-28% of B cohort fathers and 23-33% of K cohort fathers said they preferred a "little more" involvement. In other words, 62-73% of fathers in the B cohort and 72-76% of those in the K cohort wanted increased involvement in their child's life. Almost all other non-resident fathers said that their level of involvement was "about right" (B cohort: 27-36%; K cohort: 22-28%). Only 0-2% of B cohort fathers and 0-3% of K cohort fathers indicated that they would prefer less involvement. Parents' preferences regarding paternal involvement are likely to be affected by a range of factors, for example, previous and current levels of paternal involvement, the distance between the residences of father and child, and the quality of the inter-parental relationship. These factors are examined later in this chapter.

It is worth noting that the extent to which resident mothers and non-resident fathers in the LSAC study expressed preferences for greater paternal involvement seems high when compared with the results of a study by Parkinson and Smyth (2004). However, this apparent discrepancy may have resulted from differences in the measures used and the samples. For example, differences were apparent in the questions on involvement (with the Parkinson and Smyth study focusing on father-child time), the ages of the children, and the duration of parental separation. Furthermore, the preference for an increase in paternal involvement apparent in the LSAC study is consistent with the increasing recognition that the father's involvement with their children is beneficial to child development.

Preferences of partners in the sample of paired cases

The above trends focus on all resident mothers and all non-resident fathers, taken separately. As already noted, fathers who had not seen the child in the past 12 months were not interviewed. Where both parents of the same child participated in the study, it was possible to compare their preferences regarding paternal involvement from Wave 2 onwards. The patterns of preferences are divided into three broad categories as show in Table 9.2: (a) a higher level of involvement reported by the father than the mother (b) the father and mother had the same level of involvement; or (c) a lower level of involvement reported by the father than the mother.

Where the views of each parent differed, the amount of disparity was identified as represented either a difference of one rating point (e.g., "a little more involved" vs "a lot more involved" and "a little more involved" vs "current level is about right") or at least two rating points (e.g., "a lot more involved" vs "about right" and "a little more involved" vs "a little less involved"). Table 9.2 shows how each specific group was classified.

Table 9.2: Defining categories comparing mothers' and fathers' views regarding the non-resident father's involvement in their child's life, paired cases
Father's report Mother's report
A lot more involved A little more involved About right A little less involved Much less involved
A lot more involved Same Dad's > mum's, difference by one rating point Dad's > mum's, difference by at least two rating points
A little more involved Dad's < mum's, difference by one rating point Same Dad's > mum's, difference by one rating point Dad's > mum's, difference by at least two rating points
About right Dad's < mum's, difference by at least two rating points Dad's < mum's, difference by one rating point Same Dad's > mum's, difference by one rating point Dad's > mum's, difference by at least two rating points
A little less involved Dad's < mum's, difference by at least two rating points Dad's < mum's, difference by one rating point Same Dad's > mum's, difference by one rating point
Much less involved Dad's < mum's, difference by at least two rating points Dad's < mum's, difference by one rating point Same

Note: The symbol "<" means "less than" and the symbol ">" means "greater than"; for example, "Dad's > mum's" refers to cases where the father preferred greater paternal involvement than the mother.

Table 9.3 summarises the results of this comparison. As the last row in Table 9.3 shows, the number of paired cases (where both parents of the study child were interviewed) increased with each subsequent study wave. Of these three sets of responses, the most common entailed the father preferring greater paternal involvement than the mother (B cohort: 43-46%; K cohort: 43-54%), followed by the father and mother sharing the same view (B cohort: 34-38%; K cohort: 32-38%), while the least common was the father preferring less paternal involvement than the mother (B cohort: 18-21%; K cohort: 12-19%). For 18-21% of paired cases in the B cohort and 12-19% in the K cohort, the father's preferred level of paternal involvement was lower than the mother's preferred level.

Table 9.3: Preferences of fathers and mothers regarding the non-resident father's involvement in their child's life
Preference: father's vs mother's B cohort K cohort
Wave 2 (2-3 years) (%) Wave 3 (4-5 years) (%) Wave 4 (6-7 years) (%) Wave 2 (6-7 years) (%) Wave 3 (8-10 years) (%) Wave 4 (10-11 years) (%)
Father's preferred level higher than mother's 44.8 46.2 43.2 53.2 54.1 42.7
One rating point difference (adjacent ratings) 20.3 18.3 20.9 25.7 28.9 24.8
At least two rating points apart 24.5 27.9 22.3 27.5 25.2 17.9
Father's preference same as mother's 34.1 36.0 37.6 32.7 33.5 38.3
Both - a lot more involved 9.0 13.1 14.9 12.9 13.9 12.3
Both - a little more involved 4.1 9.1 4.5 6.6 5.4 10.0
Both - about right 21.0 13.8 18.2 13.2 14.2 16.0
Father's preferred level lower than mother's 21.2 17.9 19.3 14.2 12.4 19.0
One rating point difference (adjacent ratings) 13.5 14.5 14.4 9.1 8.0 15.5
At least two rating points apart 7.7 3.4 4.9 5.1 4.4 3.5
Total 100.0 100.0 100.0 100.0 100.0 100.0
No. of paired cases 91 257 355 183 370 440

Note: Due to rounding, the sum of column percentages may not total exactly 100%.

Across the three study waves, for at least one-third of this sample, both parents shared the same preference regarding the non-resident father's level of involvement. Where both parents shared the same views, they were generally less likely to prefer that the father was "a little more involved" than to prefer that the father was "a lot more involved" or to maintain the status quo (i.e., his level of involvement was deemed to be "about right"). Among B cohort parents who shared the same view, the most common response (at least for Waves 2 and 4) was a preference to maintain the status quo (reported by 18-21% of the total sample of paired cases in Waves 2 and 4 and 14% in Wave 3). A shared preference among B cohort parents for the father to become "a lot more involved" increased somewhat from 9% in Wave 2 to 15% in Wave 4. Among the K cohort where both parents shared the same views, around the same proportions indicated that they would prefer "a lot more" paternal involvement or that the level was "about right" (reported by 12-14% and 13-16% of all paired cases respectively).

Across three waves, for over 40% of the paired cases in the B cohort, the father preferred greater paternal involvement than the mother. Of all B cohort parents in the sample where both parents were interviewed, a "moderate" difference of one rating point in this direction was apparent for 18-21%, and a difference of at least two rating points in this direction emerged for 22-28%.

For over one-half of the paired cases in the K cohort in Waves 2 and 3, and 43% in Wave 4, the father preferred greater paternal involvement than the mother. Across the study waves, 25-29% of K cohort parents in this sample differed by one rating point in this direction, and 18-28% differed by at least two rating points in this direction. It is worth noting that the proportion of paired cases with at least a two-rating-point difference in this direction fell from 28% in Wave 2 to 18% in Wave 4.

As mentioned above, the least common of the three broad scenarios entailed the father expressing a preference for a lower level of paternal involvement than the mother. Similar patterns emerged across the study waves in the proportions of paired cases providing this pattern of response. A difference of one rating point was more common than a difference of at least two rating points. For example, 14-15% of all B cohort paired cases entailed a difference of one rating point in this direction, while only 3-8% entailed a difference of at least two rating points in this direction.

9.4 The changing preferences of mothers regarding the father's involvement in their child's life

The above sets of discussion focus on all resident mothers and non-resident fathers, and then on all paired cases, who participated in either Wave 2, 3 or 4, regardless of whether they participated in any other wave. This section focuses on resident mothers who had been separated from their child's father in all four waves and who had participated in each of these waves.8 The extent and nature of change in these mothers' preferences regarding the father's level of involvement in their child's life is examined. For ease of discussion, these mothers are referred to as "continuing resident mothers".

Continuing resident mothers' responses in Wave 1 are compared with those provided in each subsequent wave and are classified into three categories, following a similar format to that adopted in the comparison of responses of each mother and father of the same child (see Table 9.2). Table 9.4 presents the proportions of continuing resident mothers in the sample whose preferences regarding paternal involvement remained the same, changed in the direction of greater involvement, and changed in the direction of lesser involvement. These three broad categories are further divided as outlined in the discussion of results below.

Table 9.4: Changing preferences of resident mothers regarding non-resident father's involvement in their child's life
Change in preferences B cohort K cohort
Wave 2 vs Wave 1 (%) Wave 3 vs Wave 1 (%) Wave 4 vs Wave 1 (%) Wave 2 vs Wave 1 (%) Wave 3 vs Wave 1 (%) Wave 4 vs Wave 1 (%)
Preferred level - increased 25.6 33.1 34.3 23.5 24.9 27.6
One rating point difference (adjacent ratings) 18.6 22.1 22.6 13.6 12.5 16.2
At least two rating points apart 7.0 11.0 11.7 9.9 12.4 11.4
Preferred level - no change 48.8 40.3 28.6 49.8 42.9 41.8
No change - a lot more 21.3 20.2 11.1 17.8 15.1 16.3
No change - a little more 8.2 7.2 5.9 7.3 4.9 5.7
No change - about right 17.2 11.9 11.6 22.7 20.2 18.0
No change - a little less or much less 2.1 1.0 0.0 2.0 2.7 1.8
Preferred level - decreased 25.6 26.6 37.1 26.9 32.2 30.8
One rating point difference (adjacent ratings) 14.2 12.8 18.2 14.8 15.7 14.5
At least two rating points apart 11.4 13.8 18.9 12.1 16.5 16.3
Total 100.0 100.0 100.0 100.0 100.0 100.0
No. of mothers 124 124 124 295 295 295

Notes: Due to rounding, the sum of column percentages may not total exactly 100%. Ages of the children - B cohort: Wave 1: 0-1 year, Wave 2: 2-3 years, Wave 3: 4-5 years, Wave 4: 6-7 years; K cohort: Wave 1: 4-5 years, Wave 2: 6-7 years, Wave 3: 8-10 years, Wave 4: 11-12 years.

It is important to note that any change in mothers' preferences may or may not have resulted from an actual change in the level of paternal involvement. It also needs to be kept in mind that the needs of children would change as they grow older, as would the needs of parents as their personal circumstances change. Links between resident mothers' preferences regarding paternal involvement and their care-time arrangements (an important element of paternal involvement) are discussed in a subsequent section.

Nearly one-half of the continuing resident mothers in the B cohort held the same preference in both Wave 1 and Wave 2 (when their child was 0-1 year, then 2-3 years). Around one-quarter of the mothers had changed their view, wanting greater paternal involvement in Wave 2 than Wave 1, and around one-quarter preferred to see lower paternal involvement in Wave 2 than Wave 1. This general pattern was also apparent for Wave 3 when compared with Wave 1, though the proportion of continuing resident mothers who had changed their views towards greater paternal involvement over this 4-year period (33%) was higher than the proportion indicating this direction of change in views over the 2-year period (26%). In addition, the proportion preferring the same level of involvement over the 4-year period was lower than that apparent over the 2-year period (40% vs 49%). Over the 6-year period (from Wave 1 to Wave 4), the proportion of mothers who held the same views had fallen to 29%, while the proportion of the mothers who changed their views towards lower paternal involvement had increased to 37%. In other words, the views of continuing resident mothers in the B cohort had become more diverse with time. This may reflect the changing circumstances among these mothers, fathers and their children, including changes in personal needs, in the actual paternal levels of involvement, and/or in the quality of the inter-parental relationship.

Among continuing resident mothers in the K cohort, one-half held the same preferences over the first two waves, and 43% at Wave 3 and 42% at Wave 4 held the same views as they held in Wave 1. Compared with their preferences in Wave 1, 24% of continuing resident mothers in the K cohort wanted more paternal involvement in Wave 2, while 27% wanted less paternal involvement. By Wave 4, 28% wanted more and 31% wanted less paternal involvement.

Four scenarios involving no change in preferences are presented in Table 9.4: a continuing preference of the father to be "a lot more" involved, "a little more" involved, "a little less" or "much less" involved, and a continuing assessment that the father's level of involvement was "about right". For both cohorts, the two most common scenarios entailed a continuing preference for "a lot more" involvement and for a continuing assessment that the level of paternal involvement was "about right".

Where B cohort mothers expressed an increase in their preferred level of paternal involvement, the increase tended to reflect a change of one rating point. For example, by Wave 4, 23% of resident mothers in the B cohort had changed their views by only one rating point towards favouring paternal involvement, whereas 12% had changed their views by at least two rating points in this direction. (This trend was less apparent among K cohort mothers whose preferred level of paternal involvement had increased.) On the other hand, when mothers changed their preferred level towards less paternal involvement, decreases of one rating point and of at least two rating points were similarly likely to occur, with the patterns being consistent for both cohorts.

Overall, for both cohorts, mothers' preferences regarding paternal involvement were more likely to change as duration of separation increased. This is not surprising given the continuing changes in children's needs as they grow up and the changes in the needs of their parents as their circumstances change during the course of separation. The changes in mothers' preferences on paternal involvement were greater for those in the B cohort than K cohort. In addition, the extent of change in either direction was similar - towards increased paternal involvement and towards decreased paternal involvement.

9.5 Barriers to further involvement reported by non-resident fathers

In Waves 2-4, non-resident fathers who indicated that they would like to have greater involvement in their child's life than they had at the time were asked: "What stops you from being more involved?".9 In Waves 2 and 3, a list of 12 possible reasons (including "other reasons") was provided and non-resident parents could nominate as many of those reasons listed as they deemed appropriate. In Wave 4, only eight possible reasons were provided, and these tended to be less specific than those listed in the previous two waves. Unlike the earlier waves, non-resident parents were asked in Wave 4 to select one response option only. Most probably, they would have selected the main barrier to increased involvement. Of all the non-resident fathers who preferred to see their involvement with their child increase, the proportions of fathers who nominated the different reasons are presented in Table 9.5 (relating to Waves 2 and 3) and Table 9.6 (relating to Wave 4).

Table 9.5: Waves 2 and 3: Non-resident fathers' reasons for not being more involved
Reasons B cohort K cohort
Wave 2 (2-3 years) (%) Wave 3 (4-5 years) (%) Wave 2 (6-7 years) (%) Wave 3 (8-9 years) (%)
The demands of your job make more frequent contact difficult 50.5 43.3 50.0 36.9
Your child's other parent does not want you to see the child more often 33.7 29.0 21.8 28.3
Child lives too far away for more frequent contact & travel too expensive 18.4 24.1 31.4 29.5
The terms of a court-ordered arrangement prevent more contact 16.3 7.7 10.6 5.3
More frequent contact would be disruptive to the child's routine 9.0 2.6 21.0 2.4
Distress to child and/or self 8.0 0.0 6.8 1.0
You are prevented by illness or injury 0.0 0.0 0.0 0.4
You do not have suitable living arrangements for a child to visit 5.2 0.4 5.4 2.1
Your new partner or family makes more frequent contact difficult 1.1 3.7 1.1 2.0
Other reasons 11.6 17.1 17.2 8.6
No. of fathers 60 195 138 285

Notes: Column sums may exceed 100% because multiple responses were allowed. Questions in Wave 2 were asked in a self-completion questionnaire while questions in Wave 3 were asked in computer-assisted telephone Interviews.

Table 9.6: Wave 4: Non-resident fathers' reasons for not being more involved
Reasons B cohort (6-7 years) (%) K cohort (10-11 years) (%)
Work commitments/demands 41.7 36.9
Parent 1 (resident parent) related 29.7 21.0
Distance/cost 19.1 22.4
Current care arrangements 6.7 10.4
Family commitments/demands 0.1 1.7
Other parent living elsewhere or child commitment 1.2 4.4
Child doesn't want to 1.5 2.7
Step-family/current partner related 0.2 0.6
Total 100.0 100.0
No. of fathers 244 319

Note: Percentages may not total exactly 100% due to rounding.

Across the waves, both B and K cohort fathers most commonly nominated job-related issues as a barrier to increased involvement, followed by barriers related to the child's mother (e.g., she did not want to see increased paternal involvement) and to the distance between the two homes and cost. These patterns were consistent across the three waves. Specifically, work-related barriers to increased involvement were reported by one-half of the B and K cohort fathers in Wave 2 and by around 40% in Waves 3 and 4, while one-fifth to one-third of fathers across waves and cohorts nominated the mother as a barrier to increased involvement. Barriers of distance and cost were reported by one-fifth in Wave 2 and by nearly one-third of fathers by Wave 4 in each cohort.

The other reasons listed were nominated by a smaller proportion of fathers, with the exception of concerns about the resulting disruptions to the child's routine: one-fifth of K cohort fathers nominated this issue as a barrier to increased involvement in Wave 2 (when their child was 6-7 years old), compared with less than one in ten fathers in the B cohort in Wave 2 (when their child was 2-3 years old). Few fathers considered this as a barrier to increased involvement in Wave 3. This specific barrier was not captured in Wave 4, but may have been a consideration among those mentioning "other parent living elsewhere or child commitment"10 and the child not wanting increased involvement (each nominated by less than 5% of B and K cohorts).11 A substantial minority of non-resident fathers in Waves 2 and 3 indicated other reasons that were not specified in the questionnaire (less than one-fifth).12

9.6 Factors linked with preferences regarding fathers' involvement in their child's life

The above section focuses on fathers' beliefs about barriers to increased involvement in their child's life. A range of reasons was offered, with work-related issues, the wishes of the child's mother, and distance or costs being the most commonly mentioned. However, information on barriers was only asked of fathers who expressed a preference for more involvement. To some extent, preferences can be compromised by competing priorities. For instance, reasons for not wanting increased involvement may have included a desire to avoid upsetting the child or the child's mother, or the father lacking suitable accommodation for the child.

In this section the focus is on objective factors linked with mothers' and fathers' preferences for increased paternal involvement (as opposed to preferences for reduced involvement or for maintenance of the status quo). The factors examined were: care-time arrangements, duration of separation (expressed in terms of the child's age at separation), physical distance between the two homes, parents' current relationship status, and whether parents were at least sometimes hostile towards each other.13 The cut-off points for care-time arrangements were based on those adopted by the Department of Human Services Child Support Program for determining how much child support should be paid. Children who spent 35-65% of nights per year with each parent are referred to as having a shared care-time arrangement.

Some of these variables are also linked with socio-economic status, for example, care-time arrangements and re-partnering after separation (see de Vaus, Gray, Qu, & Stanton, 2008; Kaspiew et al., 2009). The focus of this section is on factors linked with parents' preferences for increased paternal involvement controlling some other potentially important factors. Put another way, the strength of links between paternal involvement preferences and each factor selected was assessed, when the effects of the other selected factors were controlled for.14 The analyses were carried out separately for resident mothers and non-resident fathers for each cohort and each wave. For ease of interpretation, the results are presented as estimated percentages relating to a preference for increased involvement associated with care time and distance between the two homes. (The results relating to a preference for increased involvement and other factors examined are not shown, but discussed in this section. The results of the analysis focusing on the relationship between two variables where other factors are not controlled are not presented due to limited space). Given that the broad patterns of results were similar for mothers and fathers and any table of the precise results is unwieldy, only the results for mothers are provided.15

Characteristics associated with resident mothers' preferences for greater father involvement

Table 9.7 presents the predicted percentages of resident mothers wanting to see increased (rather than reduced or no change in) the father's level of involvement in their child's life.

Table 9.7: Estimated percentages of resident mothers preferring the father to be more involved
Wave 1 (%) Wave 2 (%) Wave 3 (%) Wave 4 (%)
Care time (time with the father)
B cohort (ref. = 14-34% nights) 45.6 40.0 34.0 44.7
Father never saw the child 72.3 * 54.5 40.5 50.5
Father saw the child daytime only 55.1 59.3 ** 60.4 *** 50.8
1-13% nights 47.0 55.5 70.6 *** 71.5 ***
Shared time (35-65% nights) a - - - 11.1 ***
K cohort (ref. = 14-34% nights) 42.0 44.6 41.0 45.4
Father never saw the child 63.9 * 60.6 58.3 56.6
Father saw the child daytime only 58.2 ** 61.2 * 58.1 * 67.6 ***
1-13% nights 57.1 * 61.6 ** 56.2 * 63.8 ***
Shared time (35-65% nights) 42.0 17.8 ** 21.2 * 19.6 ***
Distance between two homes
B cohort (ref. = < 5 km) 44.8 47.2 55.3 53.4
5-19 km 47.2 47.1 39.9 46.5
20-99 km 62.9 * 42.7 45.4 43.4
100+ km 72.7 ** 76.0 *** 52.0 48.8
Other 66.5 49.1 59.7 41.2
K cohort (ref. = < 5 km)
5-19 km 42.1 32.9 34.0 56.0
20-99 km 53.1 52.8 ** 46.7 * 48.8
100+ km 56.6 * 60.0 *** 50.3 * 58.8
Other 59.4 ** 59.7 *** 54.6 ** 47.4

Notes: Ages of the children: B cohort: Wave 1: 0-1 year; Wave 2: 2-3 years; Wave 3: 4-5 years; Wave 4: 6-7 years; K cohort: Wave 1: 4-5 years; Wave 2: 6-7 years; Wave 3: 8-10 years; Wave 4: 11-12 years. Reference groups are in brackets. a Numbers in shared time in Waves 1-3 for the B cohort were too small and they were combined with the group of 14-34% of nights. The statistical significance shown in the table represents the significance of underlying coefficients in logistic regression.* p < .05, ** p < .01, *** p < .001. Predicted probabilities were computed by setting explanatory variables at their sample means. Other variables included in the model: re-partnering status, age of study child at separation, the quality of inter-parental relationship, employment, and education.

As expected, mothers' preferences were associated with their child's care-time arrangements: in general terms, the more time the child already spent in the care of their father, the lower was the likelihood that the mother would prefer to see increased paternal involvement. For example, among the B cohort mothers in Wave 4, the probability for preferring increased paternal involvement was only 11% for the mothers whose child was in a shared care-time arrangement, 45% for those whose child spent 14-34% of nights with the father (the reference group) and 72% for those whose child spent only 1-13% of nights with the father.

The underlying coefficients for these two care-time groups (shared time and 1-13% of nights) were significantly different from the reference group (i.e., 14-34% of nights - a fairly common arrangement). Specifically, the results indicate that B cohort mothers whose child spent only 1-13% of nights with the father in Wave 4 were significantly more likely than those whose child spent 14-34% of nights with the father (and even more nights than this) to indicate a preference for increased paternal involvement. On the other hand, those with shared care-time arrangements were less likely to indicate such a preference, compared with those whose child spent 14-34% of nights with the father.

Such trends are not surprising, given that care time is a central component of involvement. In section 9.5, some of the barriers to greater involvement reported by fathers pointed to their inability to spend more time with the child. However, it is important to point out that many of the relationships were not significant. For instance, except in Wave 1, B and K cohort mothers whose child never saw the father were not significantly more likely to prefer increased paternal involvement than mothers whose child spent 14-34% of nights with the father. This is not surprising given that mothers and fathers in this situation had different issues (see Kaspiew et al., 2009).

Mixed results emerged regarding links between the mothers' preferences and the distance between the two parental homes. Among the B cohort mothers in Wave 1, mothers who lived distances of at least 20 km from their child's father were significantly more likely than those living within 5 km of the father to express a preference for increased paternal involvement. In Wave 2, only those who lived at least 100 km away from the child's father were significantly more likely to prefer increased involvement than those who lived within 5 km. But in Waves 3 and 4, mothers' preferences did not vary significantly with distance between the two homes. For K cohort mothers, those with distances of at least 20 km in Wave 1 were significantly more likely to express a preference for increased paternal involvement than the reference group (those who lived within 5 km). In Waves 2 and 3, those with distances of at least 5 km were significantly more likely to prefer increased paternal time, but in Wave 4, mothers' preferences regarding paternal time did not vary significantly according to distance between the two homes. Overall, longer distance was associated with preferring more paternal involvement in Waves 1 and 2 and this link disappeared in Waves 3 and 4.

Apart from the results for B cohort mothers in Wave 3, there was virtually no apparent association between duration of separation (here expressed in terms of the age of the child at the time of separation) and mothers' preferences regarding fathers' involvement. Among B cohort resident mothers in Wave 3, those who had separated from their child's father after the child turned 2 years of age were significantly less likely to prefer increased paternal involvement, compared with their counterparts who had separated at the birth of the child or earlier (including those who never lived together). It is worth noting that, in Wave 4, B cohort resident mothers who had separated when the child was 2-4 years old were also less likely to prefer increased paternal involvement than those who separated at the birth of the child or earlier. However, no significant differences emerged for other durations of separation (here expressed in terms of the child's age at separation). For the K cohort, the child's age at the time of separation was not significantly related to mothers' preferences regarding paternal involvement.

Although partnered mothers seemed less likely to prefer increased paternal involvement than mothers who at each wave were unpartnered ("single"), this difference was only statistically significant for B cohort mothers in Wave 4 and K cohort mothers in Wave 2. For B cohort mothers, lack of significance is likely to be attributable to the small numbers who were partnered in the first three waves. However, this is not the case for K cohort mothers.

Among the B cohort mothers, those who reported that their relationship with their child's father sometimes or more frequently entailed anger or hostility were less likely to prefer increased paternal involvement than those who said that anger or hostility rarely or never occurred. However, this pattern of results was only statistically significant in Waves 1 and 3. Among K cohort mothers, no statistical differences in wanting more paternal involvement emerged between those reporting having angry or hostile episodes with their child's father and those having no such episodes. However, the variable concerning hostility included an "other" (or "not applicable") category. This referred to mothers who said that they had no contact with the father and therefore had no opportunities to have heated exchanges. K cohort mothers who indicated that they had no contact with the father were less likely to prefer increased paternal involvement than those who never or rarely experienced the emergence of anger or hostility in the relationship. This difference was statistically significant across all four waves.

Characteristics associated with non-resident fathers' preferences for greater father involvement

As was found for mothers' views, in general, fathers with shared care time were significantly less likely to express a preference for increased involvement than those who spent 14-34% of nights caring for their child. As already noted in relation to mothers, this is not surprising given that shared care time represents a great deal of involvement in the child's life. In addition, B cohort fathers with only 1-14% of care nights were significantly more likely to indicate a preference for increased involvement than those who already spent at least 14-34% of nights with their child. However, none of the other trends relating to care time were statistically significant. For instance, fathers who only saw their child during the daytime were no more (or less) likely to prefer increased involvement in their child's life than the comparison group (with 14-34% of care nights).

Regarding the effects of distance between the two homes on involvement preferences, the only statistically significant differences emerged between fathers who lived very long and very short distances away from their child.16 As expected, K cohort fathers who reported distances of at least 100 km were significantly more likely to prefer increased involvement with their child than their counterparts who lived fewer than 5 km away. This pattern also applied to B cohort fathers in Wave 3, but not Wave 4. Again, it should be kept in mind that these results emerged when the effects of differential levels of care-time arrangements were controlled.

Fathers' preferences regarding personal involvement in their child's life did not vary significantly according to their period of separation. In addition, for most of the comparisons undertaken, fathers' preferences regarding their involvement with their child did not vary significantly according to their relationship status or according to whether their relationship with their child's mother entailed anger or hostility at least some of the time. Some exceptions emerged. Firstly, for B cohort fathers in Wave 1 and for K cohort fathers in Wave 4, those who had re-partnered were more likely to prefer increased involvement than their counterparts who had remained single. Secondly, mixed results emerged for B cohort fathers regarding the effects of anger or hostility in the inter-parental relationship and fathers' involvement preferences. For example, B cohort fathers who said that they had a hostile relationship at least sometimes in Wave 3 were more likely than those who indicated that they rarely or never had a hostile relationship to prefer increased paternal involvement, while this effect was not statistically significant in Wave 4.

9.7 Conclusions

This chapter examines the preferences of separated mothers and fathers regarding paternal involvement in their child's life. The analyses focused on the views of parents in the most common arrangement, entailing the child living with the mother for most or all of the time, along with the very small proportion of cases in which the child was in an equal care-time arrangement, and where the mother had indicated that she knew the child best (and had therefore been classified as the "primary parent"). For succinctness, the mothers are referred to as "resident mothers" and the fathers as "non-resident fathers". Firstly, the views of all such parents were examined, then they were compared with the views of fathers and mothers in the paired cases sample (where the two parents of the same study child indicated their preferences). It is important to note that any fathers who had not been in face-to-face contact with their child within 12 months were excluded from responding to the study at each wave (including the "paired cases" sample).

Around one-half of the resident mothers indicated that they would prefer to see increased paternal involvement in their child's life, and most of the others said that the amount of involvement was "about right". That is, few expressed a preference for the father to be less involved than he currently was. On the other hand, the majority of non-resident fathers expressed a preference for increased involvement in their child's life. These general patterns were apparent for both cohorts across all waves. Consistent with these trends, the paired data showed that a child's father was more likely than the child's mother to prefer increased paternal involvement.

Nevertheless, preferences change as circumstances (and priorities) change. These circumstances would include, for instance, age-related changing needs of the child, events affecting parents' needs, including changing residence and those associated with the pathways they take after separation, and the fact that levels of paternal involvement (e.g., care-time arrangements) may have changed. Changes in preferences were common among mothers who had separated before Wave 1, applying to around one-half of the B and K cohort mothers by Wave 2, and to the majority by Wave 4. Changes in preferences were equally likely to be in either direction - that is, towards a preference for the father to be more involved, or less involved, than they had wanted to see in Wave 1.

This chapter also examined the barriers to increased involvement in their child's life, as perceived by non-resident fathers who expressed a preference for increased involvement. Work commitments or demands were the most commonly mentioned barriers. Other commonly mentioned issues included a belief that the child's mother did not want to see the father have increased time with his child, the physical distance between the two homes and related cost issues. No information was collected about why mothers might not want the father to have increased time with the child. Their reasons may include some of the barriers reported by fathers (e.g., distance between the homes), as well as the child's wishes, the mother's belief that the father is inept as a parent, high inter-parental conflict, family violence and so on.

The final set of analyses threw further light on reasons behind parents' preferences. Unsurprisingly, the preferences of resident mothers and non-resident fathers were linked with their care-time arrangements - the more time that the child spent with the father, the less likely were the father and mother to express a desire for increased paternal involvement. This pattern of results was stronger for the preferences expressed by resident mothers than for those expressed by non-resident fathers. The distance between the two homes was also linked with parents' preferences: those who lived only a short distance apart were less inclined to prefer increased paternal involvement, compared with those who lived a considerable distance apart. It is important to note that this trend emerged when the effects of current care-time arrangements (which would be influenced by distance) were controlled.

There was limited evidence of a link between parents' preferences and their personal relationship status. That is, mothers' and fathers' preferences varied significantly according to this factor for only one wave for each cohort. Where such differences were apparent, re-partnered mothers were less likely than those who were single to prefer increased paternal involvement, while re-partnered fathers were more likely to prefer increased involvement. Inconsistent results emerged regarding links between having a hostile inter-parental relationship and parents' preferences regarding paternal involvement. However, resident mothers who had no contact with the father (and therefore did not have the opportunity to experience episodes of anger or hostility in their dealings with the father) were less likely to prefer increased paternal involvement, compared with those who reported that episodes of anger or hostility rarely or never emerged. Finally, parents' preferences did not vary significantly according to the time lapse since separation.

It is important to keep in mind that the number of non-resident fathers participating in the study was smaller than that of resident mothers, especially in Wave 2. The characteristics of non-resident fathers who participated may differ somewhat from those who did not and thus the results of non-resident fathers may be biased and should be interpreted with caution. In Wave 3, some resident parents were given the opportunity to skip the section on parenting issues after separation and this may introduce some bias in the results of resident mothers in this wave.

It should be pointed out that fathers' involvement is multidimensional and in expressing their preferences, parents may well focus on different aspects of paternal involvement. However, as stated above, it does seem likely that care-time arrangements were commonly considered. Whether intended or not, parents' preferences regarding fathers' involvement may affect the father-child relationship, through encouraging or discouraging face-to-face time and other ways of spending time together. Children's changing developmental needs during the course of separation are likely to come into play, as reflected in the fact that children of primary school age are considerably more likely than younger and older children to experience shared care time (Kaspiew et al., 2009). Consistent with findings in previous research, this chapter shows that a desire for increased involvement of fathers is prevalent among both resident mothers and non-resident fathers, but contingent on the existing level of care time. The 2006 amendments to the Australian family law system were intended to encourage and facilitate the continuing involvement of both parents where children's safety is not at risk. Future research will be able to tell whether the groundswell of support for both parents' involvement in children's lives after separation, encouraged by the changes to the family law system, will be a reality for more children after parental separation, and whether adequate protection is in place so that children's "best interests" are served.17

9.8 References

  • Allen, S., & Daly, K. (2007). The effects of father involvement: An updated research summary of the evidence. University of Guelph, Canada: Father Involvement Research Alliance.
  • Amato, P. R., & Gilbreth, J. (1999). Nonresident fathers and children's well-being: A meta-analysis. Journal of Marriage and the Family, 61, 557-573.
  • Australian Bureau of Statistics (2011). Family characteristics, Australia, 2009-10. (Cat. No. 4442.0). Canberra: Australian Bureau of Statistics.
  • Baxter, J., Edwards, B., & Maguire, B. (2012). New father figures and fathers who live elsewhere (Occasional Paper No. 42). Canberra: Department of Families, Housing, Community Services and Indigenous Affairs.
  • Craig, L., Mullan, K., & Blaxland, M. (2010). Parenthood, policy and work-family time in Australia. Work, Employment and Society, 24(1), 27-45.
  • de Vaus, D., Gray, M., Qu, L. & Stanton, D. (2008). The financial consequences of divorce for later life. In P. A. Kemp, K. Van den Bosch, & L. Smith (Eds.), Social protection in an ageing world (pp. 257-278). Oxford: Intersentia.
  • Flouri, E. (2007). Fathering and adolescents' psychological adjustment: The role of fathers' involvement, residence and biology status. Child: Care, Health and Development, 34(2),152-161.
  • Kaspiew, R., Gray, M., Weston, R., Moloney, L., Hand, K., Qu, L., & the Family Law Evaluation Team (2009). Evaluation of the 2006 family law reforms. Melbourne: Australian Institute of Family Studies.
  • Lamb, M.E. & Lewis, C. (2012). The development and significance of father-child relationships in two-parent families. In M.E. Lamb (Ed.), The role of the father in child development (4th ed., pp. 272-306). New Jersey: John Wiley & Sons, Inc.
  • Marsiglio, W., Amato, P., Day, R. D., & Lamb, M. E. (2000). Scholarship on fatherhood in the 1990s and beyond. Journal of Marriage and the Family, 62, 1173-1191.
  • Moloney, L., Weston, R., & Hayes, A. (2013). Key social issues in the development of Australian family law: Research and its impact on policy and practice. Journal of Family Studies, 19(2), 110-138.
  • Parkinson, P., & Smyth, B. (2004). Satisfaction and dissatisfaction with father-child contact arrangements in Australia. Child and Family Law Quarterly, 16, 289-304.
  • Qu, L., & Weston, R. (2010). Parenting dynamics after separation: A follow-up study of parents who separated after the 2006 family law reforms. Melbourne: Australian Institute of Family Studies.
  • Renda, J. (2013). Is it just a matter of time? How relationships between children and their separated parents differ by care-time arrangements. In Australian Institute of Family Studies, The Longitudinal Study of Australian Children Annual Statistical Report 2012. Melbourne: Australian Institute of Family Studies.

Endnotes

1 Not all studies have observed positive links between paternal involvement and children's developmental progress. Amato and Gilbreth (1999), for instance, found no significant relationship between frequency of contact per se and children's developmental progress. However, depending on the way it is measured, frequency of contact may not measure the overall amount of time fathers spend with their children. For instance, fortnightly contact may entail one or several overnight stays per fortnight.

2 Subsequent to these reforms, the Family Law Legislation (Family Violence and other Measures) Act 2011 (Family Violence Act) was introduced in order to increase the protection of children and other family members from family violence or child abuse. The Act came in to operation in June 2012.

3 Most fathers who expressed safety concerns indicated that these related to their child alone, whereas mothers were almost equally likely to indicate that they held safety concerns for their child alone or for both their child and self.

4 Parents living elsewhere were not recruited for the study if they had not seen the study child at least once a year.

5 The Department of Human Services Child Support Program considers children to be in a shared care-time arrangement if they spend 35-65% of nights with each parent in a year. This classification has been adopted for this study.

6 Of the study children with a parent living elsewhere, only a small proportion (1-3% across four waves for either cohort) were either in equal time (i.e., entailing 48-52% of time with each parent) or shared time entailing more nights with the father than mother (i.e., 53-65% of nights per year with the father and 35-47% of nights with the mother).

7 A small number of separated parents reconciled and thus were out of the scope in later waves.

8 Due to a small number of non-resident fathers who were interviewed in all waves, the analysis was not carried out for a continuing sample of non-resident fathers.

9 Baxter et al. (2012) also examined the barriers to increased involvement reported in Wave 3 by non-resident fathers who wanted increased involvement in their child's life. This section expands on their analysis by including fathers' perceptions on this issue collected in Waves 2 and 4 (as well as Wave 3).

10 This may refer to a new partner or another child (e.g., step-child or child born of another relationship, or the study child's full sibling who spends most of the time in the care of the father).

11 In Wave 2, 5% of B cohort fathers nominated the child's distress associated with "change-overs" in care time as a barrier to increased involvement. The children in this cohort were 2-3 years old at the time. No B cohort fathers mentioned this issue in Wave 3 (when their child was 4-5) and only around 1% of K cohort fathers mentioned this issue in Waves 2 and 3 (when their child was 6-7 years and 8-9 years).

12 The response option "Other reasons" was not provided in Wave 4.

13 In each wave, resident parents were asked, "How often is there anger or hostility between you and the child's other parent?" Responses to this question are divided into three groups: (a) never or rarely; (b) sometimes, often or always; and (c) other (i.e., no contact between the parents, not applicable).

14 Socio-economic characteristics were measured by two variables: employment status and level of education. Some of the factors examined would have been inter-related. For example, distance between the two parental homes tends to be associated with children's care-time arrangements: the closer their location, the more likely the children are to spend substantial time with their non-resident parent or to have a shared care-time arrangement (Kaspiew et al., 2009). Nevertheless, there is no collinearity across the variables included in the regressions.

15 The results pertaining to fathers' preferences are available on request.

16 That is, their child's usual residence.

17 The introduction of the Family Violence Act, which came into operation in June 2012 is relevant here. It was designed to increase the protection of children and other family members from family violence or abuse.

Top